In this proposed study a quantitative evaluation of the dynamic exercise hyperpnea will be performed in normal, healthy individuals to ascertain the significance of the neural and humoral components of this process. Stochastic variations in the applied exercise demands are to be presented. Deterministic signals previously utilized to study exercise have the disadvantage of subjective interference from anticipation of the stimulus (e.g., sinusoidals and ramps), generating a startle response (e.g., short duration pulses and steps), or not adequately stimulating the faster dynamic (fast neural) components. Stochastic exercise stimuli applied in this study will be optimally chosen to minimize the anticipatory interference and startle component of the spontaneous ventilatory response to dynamic exercise, and to adequately stimulate the higher frequency components of this response. In addition, the hypoxic augmentation of the dynamic exercise hyperpnea will be quantitatively investigated, and the exercise augmentation of the dynamic CO2 sensitivity will be evaluated for its contribution to the exercise hyperpnea. Standard signal processing procedures of spectral density and correlation techniques provide a basis for quantitative evaluation of these contributions in both the frequency and time domains while minimizing noise components in the signal uncorrelated with the stimulus. The analysis of the results of this study will provide a quantitative evaluation of the humoral and neural components of exercise hyperpnea and the relation of these components to the control of ventilation. Although not planned as a part of this study, the procedures developed here have an avenue for direct application to patient evaluation.